Understanding the RPI, 102: The Real Estate Rule

Stop. If you haven’t read part one of this case study in idiocy, please click the link now. Think of it as a prerequisite that you should at least skim before moving on to the next chapter. But just in case you need a refresher, here’s a 30,000 foot summary of URPI 101:

A) Half of a team’s RPI is determined by its opponents’ winning percentage

2) Shakespeare would not like the RPI

D) Losing a game can be better than winning a game by 42

Garnet) Playing a crappy team that’s in an even crappier conference is better than playing a crappy team in a really good conference

The Real Estate Rule

Alright, alright, alright. Welcome to URPI 102, featuring the “real estate” rule. If you recall, in URPI 101 I mentioned the presence of several nuances within the three primary factors. Perhaps the most significant of these nuances—at least as it relates to impact on the actual index (and ultimately on bubbles around the country)—is the real estate rule.

The real estate rule is all about, you guessed it, location, location, location. It goes like this:

• Wins on your home court are weighted at 0.6.
• Wins on a neutral court are weighted at 1.0.
• Wins on an opponents’ home court are weighted at 1.4.
• Losses are the reverse. 1.4 for a home loss; 1.0 for a neutral court loss; 0.6 for a road loss.

The Missouri Valley House of Cards

The RPI did not always contain this nuance. Sometime around the year 2000, people started complaining that mid and low-majors were at a disadvantage when it came to Selection Sunday, in part because they had to play so many road games in the non-conference portion of the season in order to get a shot at the big dogs. And road games are indeed harder to win than home games.

Before the 2004-2005 season, the RPI formula was changed to include the nuance detailed above and one conference in particular took notice. The Missouri Valley secured two NCAAT bids (the tourney champ and one at large) the five previous seasons. But in the 2005 season the conference broke through and earned 3 bids. Then in 2006, the MVC earned 4 bids—equaling the amount handed out to the Pac-10, Big 12, and ACC. And they did it by scheduling road games.

In 2004, the top 5 teams in the MVC played a total of 11 OOC road games. In 2005, the first year of the new formula, the top 5 teams in the league scheduled 15 OOC road games. That’s nearly an extra road game per team. In 2006, the number jumps to 19 OOC road games. From roughly 2 per team in 2004 to almost 4 per team by 2006. The result was akin to a house of cards.

Recall the three factors of the RPI. Factor number one dictates that 25% of your team’s RPI is determined by your team’s winning percentage. Cue the real estate rule. If Team A plays 10 home games and wins 8, us humans see them as 8-2. Which is of course a 0.8 winning percentage. However, the RPI does not see 8-2. To the RPI, Team A has a record of 4.8-2.8...which is only a .631 winning percentage.

Same team. Same opponents. Same results. Significantly different winning percentage.

Of course, the opposite effect happens if you play road games. If Team B plays 10 road games and goes 5-5 in these games, humans see Team B as having a 0.5 winning percentage. But the RPI sees Team B’s record as 7-3, good for a 0.7 winning percentage.

That look on your face...it’s one I’m familiar with. I know what you’re thinking. How could a team with an 8-2 record have a lower winning percentage than a team with a 5-5 record? In what world world does that make sense? Now you know why it’s a case study in idiocy.

Back to the MVC house of cards. By playing nearly twice as many road games, Missouri Valley teams literally won more games (or lost fewer games) to the RPI ratings. This increase in winning percentage allowed MVC teams higher RPIs than they otherwise would have enjoyed. Which meant most MVC games featured two top-100 RPI teams, and RPI top-100 wins look quite good on a bubble team’s resume. At the same time, it meant very few conference losses resulted in a sub-150 loss—which is a red flag to committee members.

Come Selection Sunday, Northern Iowa, an 11-7 team in the MVC (the 7th best league in the country in 2006), earned an at-large bid as a 10 seed. Meanwhile, a 9-7 team from the ACC (the 2nd best conference in the country that season) was relegated to the NIT. That team was Florida State.

A Modern Exemplar

More than a decade after the MVC House of Cards, there’s another example of gaming the Real Estate rule to perfection, and it’s courtesy of the reptiles southeast of Tallahassee. The University of Florida just completed a sparkling new renovation to the “O Dome”. Intended to be a recruiting benefit, the construction is having a direct benefit on the Gators’ NCAA Tournament hopes this season.

On December 21, UF played their first true home game of the 2016-17 basketball season. Prior to that win over Arkansas Little-Rock, UF had played nine neutral-site games and two road games. Accordingly, the Gators’ 9-3 real life record is 9-2.6 in RPI land. By comparison, FSU’s 12-1 real life record is 8.4-1.0 in the world of RPI.

These games away from home (combined with UF playing a strong slate of opponents) have the Gators sitting pretty with a top-eight RPI heading into conference play. And given the overall weakness of the SEC, Florida’s essentially a lock for an at-large bid in March. In fact, even a 9-9 conference mark would leave UF with a projected RPI around 35. That’s what I call constructing a strong schedule.

In the next installment of this series, we’ll examine the old adage of what’s good for the goose is good for the gander.